Please Don't Say it's Food! Nutrients and Breast Cancer Risk
Studies trying to pin breast cancer risk to intake of various food groups have been to say the least, inconsistent. The authors performed a nutrient-wide association study (NWAS), not dissimilar to the genome-wide association study (GWAS). They separately estimated associations for each food and nutrient and selected promising associations in an independent study. Two large databases were used. The NLCS included 120,852 participants about half of which were 55-69 when recruited in the Netherlands. The EPIC study included 521,330 men and women accrued between 1992-2000.
After various exclusions in the EPIC study, 272,098 women were available or study. There were 10,979 invasive breast cancers during the median follow-up time of 15 years. Of the 92 foods and nutrients evaluated, 6 were associated with impacting the risk of breast cancer. Higher intakes of wine, alcohol, beer, and cider were associated with a higher breast cancer risk (R=1.05, 955 CI 1.03-1.06). On the other hand, higher intakes of fiber/pear and carbohydrates were associated with a lower risk of breast cancer (R=0.96 95% CI 0.96-1.0). In a separate analysis, only postmenopausal women were associated with a higher risk for alcohol, cider, wine and beer and lower risk with fiber intake.
The magnitude and direction in the NLCS were similar to each factor in the EPIC study, confirming the findings of that study. The authors note that “This study reaffirms the positive association between alcohol and postmenopausal breast cancer risk.” It was noted that when considering the sources of fiber, the inverse association was driven largely by fiber from vegetables and fruit but not fiber from cereal or other sources. In the Million Women Study in the UK, a reduction in risk of breast cancer was also noted in postmenopausal women ingesting fruit but not vegetables. A meta-analysis of 16 prospective studies further confuses the topic when noting a decreased risk for women ingesting large quantities of soluble vs insoluble fiber. The inverse association of risk with apple/pear may reflect the general tendency toward fiber intake. Furthermore, in a meta-analysis of 10 prospective studies, higher fruit intake was associated with a higher risk of breast cancer.
Overall, the fact that few foods and nutrients were found to be associated with breast cancer risk may reflect that diet in middle-aged or recent diet likely does not play an important role in breast cancer development. Small as they are the associations in these studies do have a plausible biologic mechanism. For instance, fiber may inhibit estrogen absorption lowering circulating estrogens. Alcohol has been shown to increase the level of circulating estrogens. Nonetheless, the mechanism by which alcohol increases breast cancer risk if poorly understood. These 2 studies confirm the established increase in risk associated with alcohol consumption.
The authors conclude “That this study confirms the well-established increase in the risk of breast cancer associated with alcohol consumption and suggests a higher intake of dietary fiber and possibly fruit might be associated with reduced breast cancer risk.”
EDITORIAL COMMENT: I read this paper with interest considering the large number of subjects. Despite the statically significant numbers one might tend to be generally underwhelmed by the small increases in relative risk with HRs of 1.05-1.07 and decreases to 0.96. However, with the enormity of the number of women developing breast cancer worldwide each year, even a small decrease in relative risk may have a major impact on the number of yearly breast cancer cases.